ES2449623T3 - Bispecific antigen binding proteins - Google Patents

Abstract

Bispecific antigen binding protein, comprising: a) two light chains and two heavy chains of an antibody, which specifically bind to a first antigen and comprising two Fab fragments, b) two additional Fab fragments of a specifically binding antibody to a second antigen, where said additional Fab fragments are fused, both by means of a linker peptide, with the chains weighed in a), where, in the Fab fragments the following modifications are carried out: i) in both Fab fragments of a) , or in both Fab fragments of b), the variable domains VL and VH substitute one for the other, and / or the constant domains CL and CH1 are substituted one for the other, ii) in both Fab fragments in a), the variable domains VL and VH are substituted for each other, and the constant domains CL and CH1 are mutually substituted, and in both Fab fragments in b), the variable domains VL and VH are substituted for each other, or the constant domains CL and CH1 are substituted one by another, iii) in both Fab fragments in a) the variable domains VL and VH are substituted one by one, the constant domains CL and CH1 are mutually substituted, and in both Fab fragments in b), the variable domains VL and VH are they substitute one for the other, and the constant domains CL and CH1 are substituted one for the other, iv) in both Fab fragments in a), the variable domains VL and VH are replaced one by another, and in both Fab fragments in b), the domains CL and CH1 constants substitute each other, or v) in both Fab fragments in a), the constant CL and CH1 domains substitute each other, and in both Fab fragments in b), the variable domains VL and VH substitute each other.

Description

Bispecific antigen binding proteins

The present invention relates to bispecific antigen-binding proteins, methods for their production, pharmaceutical compositions containing said antibodies and uses thereof.

Background of the invention

Manipulated proteins, such as bispecific or multispecific antibodies capable of binding to two or more antigens, are known in the art. These multispecific binding proteins can be generated by cell fusion, chemical conjugation or recombinant DNA techniques.

Recently a wide variety of recombinant multispecific antibody formats have been developed, for example tetravalent bispecific antibodies by fusion of, for example, an IgG antibody format and single chain domains (see, for example, Coloma MJ et al., Nature Biotech 15: 159-163, 1997; WO No. 2001/077342, and Morrison SL, Nature Biotech. 25 (2007) 1233-1234.

In addition, other new formats have been developed in which the structure of the antibody core (IgA, IgD, IgE, IgG or IgM) is no longer conserved, such as diabodies, triabodies or tetrabodies, minibodies, various single chain formats (scFv , Bis-scFv), which are capable of binding to two or more antigens (Holliger P. et al., Nature Biotech. 23: 1126-1136, 2005; Fischer N. and Léger O., Pathobiology 74: 3-14, 2007; Shen J. et al., J. Immunol. Methods 318: 65-74, 2007; Wu C. et al., Nature Biotech. 25: 1290-1297, 2007).

All such formats use connectors, to fuse the antibody core (IgA, IgD, IgE, IgG or IgM) to an additional binding protein (eg scFv) or to fuse, for example, two Fab or scFv fragments (Fischer N. and Léger O., Pathobiology 74: 3-14, 2007). Although it is clear that the connectors have advantages for the manipulation of bispecific antibodies, they can also cause problems in therapeutic contexts. Indeed, these foreign peptides could induce an immune response against the same poker

or against the union between the protein and the poker. In addition, the flexible nature of these peptides causes them to have a greater tendency to proteolytic cleavage, potentially leading to poor antibody stability, aggregation and increased immunogenicity. In addition, it may be desired to retain effector functions, such as, for example, complement-dependent cytotoxicity (CDC) or antibody-dependent cellular cytotoxicity (ADCC), which are mediated by the Fc part by maintaining a high degree of similarity. to natural antibodies.

Thus, ideally the objective should be the development of bispecific antibodies that are very similar in general structure to natural antibodies (eg IgA, IgD, IgE, IgG or IgM), with minimal differences with respect to human sequences.

In one approach, bispecific antibodies have been produced that are very similar to natural antibodies using quadroma technology (see Milstein C. and Neck AC, Nature 305: 537-540, 1983) based on the somatic fusion of two hybridoma cell lines different expressing murine monoclonal antibodies with the desired specificities of the bispecific antibody. Due to the random pairing of the heavy and light chains of two different antibodies within the resulting hybrid hybridoma (or quadroma) cell line, up to ten different species of antibody are generated, of which only one is the desired functional bispecific antibody. Due to the presence of poorly matched secondary products, and significantly reduced production yields, sophisticated purification procedures are necessary (see, for example, Morrison S.L., Nature Biotech. 25 (2007) 1233-1234). In general, the same problem of poorly matched by-products is maintained in case recombinant expression techniques are used.

An approach to avoid the problem of poorly matched by-products, known as the "buttonhole button," is intended to force the matching of two different heavy chains of antibodies by introducing mutations in the CH3 domains to modify the contact interface. In a chain, bulky amino acids are replaced by amino acids with short side chains, to create an “eyelet”. Conversely, amino acids with large side chains are introduced into the other CH3 domain, to create a "button." Through the coexpression of these two heavy chains (and two identical light chains, which must be appropriate for both heavy chains), high yields of heterodimer formation ("buttonhole") have been observed against the formation of homodimers (" buttonhole ”or“ button-button ”) (Ridgway JB, Protein Eng. 9: 617-621, 1996, and WO No. 96/027011). The percentage of heterodimers could be further increased by remodeling the interaction surfaces of the CH3 domains using a phasic expression approach and the introduction of a disulfide bridge to stabilize the heterodimers (Merchant AM et al., Nature Biotech. 16: 677681, 1998; Atwell S. et al., J. Mol. Biol. 270: 26-35, 1997). New approaches to technology are described

button-buttonhole in, for example, EP No. 1870459A1. Although this format appears to be very attractive, there are currently no data describing the progression towards clinical use. An important limitation of this strategy is that the light chains of the two parental antibodies must be identical to avoid incorrect pairing and the formation of inactive molecules. Thus, this technique is not appropriate for the simple development of recombinant bispecific antibodies against two antigens starting from two antibodies against the first and second antigens, because the heavy chains of these antibodies and / or identical light chains must be optimized.

Another way to avoid the problem of the appearance of incorrectly paired secondary products during the preparation of bispecific antibodies is the use of heterodimers instead of homodimers, by using a full-length antibody that specifically binds to a first antigen and to which two fused Fab fragments that specifically bind to a second antigen have been fused at the N-terminal ends of their heavy chains, as described in, for example, WO No. 2001/077342. A major disadvantage of this strategy is the formation of unwanted inactive byproducts due to the incorrect pairing of the full-length antibody light chains with the CH1-VH domains of the Fab fragments and the incorrect pairing of the Fab fragment light chains. with the CH1-VH domains of the full length antibody.

WO 2006/093794 refers to heterodimeric protein binding compositions. WO 99/37791 describes antibody derivatives for multiple purposes. Morrison S.L. et al., J. Immunolog. 160: 28022808, 1998, refer to the influence of domain exchange in the variable region on the functional properties of IgG.

Summary Description of the Invention

The invention comprises a bispecific antigen binding protein, comprising:

a) two light chains and two heavy chains of an antibody, that specifically bind to a first antigen and that comprise two Fab fragments, b) two additional Fab fragments of an antibody, which specifically bind to a second antigen, in wherein said additional Fab fragments are fused, both by a linker peptide, to the C-terminal or N-terminal ends of the heavy chains or light chains in a), and where, in the Fab fragments the following modifications are carried out:

i) in both Fab fragments of a), or in both Fab fragments of b), the variable domains VL and VH are replaced by one, and / or the constant domains CL and CH1 are replaced by one,

ii) in both Fab fragments of a) the variable domains VL and VH are substituted one by one, and the constant domains CL and CH1 are substituted one by another, and in both Fab fragments of b) the variable domains VL and VH are replaced one by another, or the constant domains CL and CH1 are replaced one by another

iii) in both Fab fragments of a) the variable domains VL and VH are substituted one by one, the constant domains CL and CH1 are substituted one by one, and in both Fab fragments of b) the variable domains VL and VH are substituted one on the other, and the constant domains CL and CH1 are replaced by one,

iv) in both Fab fragments of a) the variable domains VL and VH are replaced by one, and in both Fab fragments of b) the constant domains CL and CH1 are replaced by one, or

v) in both Fab fragments of a) the constant domains CL and CH1 are substituted for one another, and in both Fab fragments of b) the variable domains VL and VH are substituted for each other.

A further embodiment of the invention is a method for the preparation of an antigen binding protein according to the invention, comprising the steps of:

a) transform a host cell with

-

vectors comprising nucleic acid molecules encoding a bispecific antigen-binding protein according to the invention

b) culturing the host cell under conditions that allow the synthesis of said antibody molecule, and

c) recovering said antibody molecule from said culture.

A further embodiment of the invention is a host cell comprising:

-

vectors comprising nucleic acid molecules encoding an antigen binding protein 5 according to the invention.

A further embodiment of the invention is a pharmaceutical composition comprising an antigen binding protein according to the invention and at least one pharmaceutically acceptable excipient.

A further embodiment of the invention is a method for the treatment of a patient in need of therapy, characterized by the administration in the patient of a therapeutically effective amount of an antigen binding protein according to the invention.

According to the invention, the ratio between a bispecific antigen binding protein desired and

15 unwanted by-products, by substituting certain domains a) in the Fab fragment of the full-length antibody that specifically binds to a first antigen and / or b) in the two additional fused Fab fragments. In this way, the unwanted mismatch of light chains with the incorrect CH1-VH domains can be reduced.

20 Detailed Description of the Invention

The invention comprises a bispecific antigen binding protein, comprising:

a) two light chains and two heavy chains of an antibody, which specifically bind to a first antigen and comprising two Fab fragments,

b) two additional Fab fragments of an antibody, which specifically bind to a second antigen, in which said additional Fab fragments are fused, both by a linker peptide, with the C-terminal or N-terminal ends of the heavy chains in a) , and in which Fab fragments were carried out

30 the following modifications:

i) in both Fab fragments of a), or in both Fab fragments of b), the variable domains VL and VH are replaced by one, and / or the constant domains CL and CH1 are replaced by one,

Iii) in both Fab fragments of a) the variable domains VL and VH are substituted one by one, and the constant domains CL and CH1 are replaced one by another, and in both Fab fragments of b) the variable domains VL and VH are substituted one on the other, the constant domains CL and CH1 are replaced by one,

Iii) in both Fab fragments of a) the variable domains VL and VH are substituted one by one, the constant domains CL and CH1 are substituted one by one, and in both Fab fragments of b) the variable domains VL and VH are replaced one for another, and the constant domains CL and CH1 are replaced one by another,

iv) in both Fab fragments of a) the variable domains VL and VH are replaced by one, and in both 45 Fab fragments of b) the constant domains CL and CH1 are substituted by one another, or

v) in both Fab fragments of a) the constant domains CL and CH1 are substituted for one another, and in both Fab fragments of b) the variable domains VL and VH are substituted for each other.

In one embodiment of the invention, the bispecific antigen-binding protein according to the invention is characterized in that said additional Fab fragments are fused, both by means of a linker peptide, to the Final ends of the heavy chains in a), or to the N-ends. -determinals of heavy chains in a).

In one embodiment of the invention, the bispecific antigen binding protein according to the invention is characterized

55 because said additional Fab fragments are fused, both by means of a linker peptide, to the Final ends of the heavy chains in a).

In one embodiment of the invention, the bispecific antigen-binding protein according to the invention is characterized in that said additional Fab fragments are fused, both by means of a linker peptide, to the N-termini

60 heavy chain terminals in a).

In one embodiment of the invention, the bispecific antigen-binding protein according to the invention is characterized in that the following modifications were carried out in the Fab fragments:

i) in both Fab fragments of a), or in both Fab fragments of b), the variable domains VL and VH are substituted for one another, and / or the constant domains CL and CH1 are mutually substituted.

In one embodiment of the invention, the bispecific antigen-binding protein according to the invention is characterized in that the following modifications were carried out in the Fab fragments:

i) in both Fab fragments of a) the variable domains VL and VH are substituted for one another, and / or constant domains CL and CH1 are substituted for each other.

In one embodiment of the invention, the bispecific antigen-binding protein according to the invention is characterized in that the following modifications were carried out in the Fab fragments:

i) in both Fab fragments in a)

the constant domains CL and CH1 substitute each other.

In one embodiment of the invention, the bispecific antigen-binding protein according to the invention is characterized in that the following modifications were carried out in the Fab fragments:

i) in both Fab fragments of b) the variable domains VL and VH are substituted for one another, and / or constant domains CL and CH1 are substituted for each other.

In one embodiment of the invention, the bispecific antigen-binding protein according to the invention is characterized in that the following modifications were carried out in the Fab fragments:

i) in both Fab fragments in b)

the constant domains CL and CH1 substitute each other.

According to the invention, the proportion between bispecific antigen-binding protein and unwanted by-products (due to incorrect pairing of light chains with the incorrect CH1-VH domains) can be reduced by replacing certain domains a) in the fragment Fab of the full length antibody that specifically binds to the first antigen and / or b) in the two additional fused Fab fragments. Incorrect pairing in the present context refers to the association of: i) the full length antibody light chain in a) with the CH1-VH domains of the Fab fragments in b), or ii) the light chain of the fragments Fab in b) with the CH1-VH domains of the full length antibody in a) (see Fig. 3), leading to unwanted inactive or non-fully functional side products.

The term "antibody" as used herein refers to a full-length antibody consisting of two antibody heavy chains and two antibody light chains (see Fig. 1). A heavy chain of a full length antibody is a polypeptide consisting of the N-terminal to C-terminal direction of an antibody heavy chain variable domain (VH), a heavy chain constant domain (CH1) of antibody, a hinge region (HR) of antibody, a heavy domain constant domain 2 (CH2) of antibody, and a heavy domain constant domain 3 (CH3) of antibody, abbreviated: VH-CH1-HR-CH2-CH3, and optionally a constant heavy chain (CH4) antibody domain 4 in the case of an antibody of the IgE subclass. Preferably, the heavy chain of the full length antibody is a polypeptide consisting, in the N-terminal to C-terminal direction, of VH, CH1, HR, CH2 and CH3. The full-length antibody light chain is a polypeptide consisting, in the N-terminal to C-terminal direction, of an antibody light chain (VL) variable domain, and an antibody light chain (CL) constant domain, abbreviated: VL-

CL. The light chain constant domain (CL) of antibody can be K (kappa) or A (lambda). The antibody chains are linked together by disulfide bonds between polypeptides, between the CL domain and the CH1 domain (ie, between the light and heavy chains and between the hinge regions of the heavy chains of the full length antibody. Examples of typical full length antibodies are natural antibodies such as IgG (for example IgG1 and IgG2), IgM, IgA, IgD and IgE.The antibodies according to the invention can be of a single species, for example human, or they can be chimerized antibodies or humanized, the full length antibodies according to the invention comprise two antigen binding sites, each formed by a pair of VH and VL, both specifically binding the same (first) antigen, the C-terminal end of the heavy chains and light of said full length antibody indicates the last amino acid at the Cterminal end of said heavy or light chain Said antibody comprises two fragments Identical Fabs consisting of the VH and CH1 domains of the heavy chain and the VL and CL domains of the light chain (see figs. 1 and 2).

An "additional Fab fragment" (see Fig. 2) of an antibody that specifically binds to a second antigen refers to an additional Fab fragment consisting of the VH and CH1 domains of the heavy chain and the VL and CL domains of the light chain of said second antibody. Additional Fab fragments are fused in their unmodified form (see Fig. 3) by the heavy chain part (the CH1 or VH domain) to the C-terminal or N-terminal ends of the light or heavy chains of the antibody that specifically binds to a first antigen.

The term "linker peptide" as used in the invention refers to a peptide with amino acid sequences that is preferably of synthetic origin. These connecting peptides according to the invention are used to fuse the antigen-binding peptides to the C-terminal or N-terminal end of the modified full-length and / or full-length antibody chains, to form a bispecific antigen-binding protein. according to the invention. Preferably, said connecting peptides under c) are peptides with an amino acid sequence having a length of at least 5 amino acids, preferably with a length between 5 and 100, more preferably between 10 and 50 amino acids. In one embodiment, said connecting peptide is (GxS) not (GxS) nGm, with G = glycine, S = serine and (x = 3, n = 3, 4, 5 or 6, and m = 0, 1, 2 or 3 ) or (x = 4, n = 2, 3, 4 or 5, and m = 0, 1, 2 or 3), preferably x = 4 and n = 2 or 3, more preferably with x = 4 and n = 2. In one embodiment, said connecting peptide is (G4S) 2.

The terms "binding site" or "antigen binding site" as used herein refers to one or more regions of an antigen binding protein according to the invention to which a ligand binds (for example the antigen or a fragment thereof) and that is derived from an antibody molecule or a fragment thereof (for example a Fab fragment). The antigen binding site according to the invention comprises variable heavy chain (VH) antibody domains and variable light chain (VL) antibody domains.

Antigen binding sites (i.e., VH / VL couples) that specifically bind to the desired antigen can be derived a) from known antigen antibodies, or b) from new antibodies or antibody fragments obtained by de novo immunization methods. using, among others, the antigen protein or nucleic acid, or fragments thereof, or by phageal expression.

An antigen binding site of an antigen binding protein of the invention contains six complementarity determining regions (CDRs) that contribute in varying degrees to the affinity of the antigen binding site. There are three heavy chain variable domain CDRs (CDRH1, CDRH2 and CDRH3) and three light chain variable domain CDRs (CDRL1, CDRL2 and CDRL3). The extent of the CDR and frame regions (FRs) is determined by comparison with a database compiled from amino acid sequences in which said regions have been defined according to the variability between sequences.

Antibody specificity refers to the selective recognition of the antibody or antigen binding protein for a particular epitope of an antigen. Natural antibodies, for example, are monospecific. Bispecific antibodies are antibodies that have two different antigen binding specificities. In the event that an antibody has more than one specificity, the recognized epitopes can be associated with a single antigen or more than one antigen.

The term "monospecific" refers to an antigen-binding antibody or protein as used herein refers to an antigen-binding antibody or protein that has one or more binding sites, each of which binds. to the same epitope of the same antigen.

The term "valid" as used in the present application refers to the presence of a specific number of binding sites in an antibody molecule. A natural antibody, for example, or a full length antibody according to the invention, has two binding sites and is bivalent. The term "tetravalent" refers to the presence of four binding sites in an antigen binding protein. The term "tetravalent bispecific" as used herein refers to an antigen binding protein according to the invention that has four antigen binding sites from which two bind to another antigen (or another antigen epitope ). The antigen binding proteins of the present invention have four binding sites and are tetravalent.

The full length antibodies of the invention comprise immunoglobulin constant regions of one or more classes of immunoglobulin. Immunoglobulin classes include the IgG, IgM, IgA, IgD and IgE isotypes and, in the case of IgA and IgA, the subtypes thereof. In a preferred embodiment, a full length antibody of the invention has a constant domain structure of an IgG antibody.

The terms "monoclonal antibody" or "monoclonal antibody composition" as used herein refers to an antibody or antibody preparation or antigen-binding protein molecules of a single amino acid composition.

The term "chimeric antibody" refers to an antibody that comprises a variable region, that is the binding region, from a source or species, and at least a part of a constant region derived from a different source or species, usually prepared by recombinant DNA techniques. Chimeric antibodies comprising a murine variable region and a human constant region are preferred. Other preferred forms of "chimeric antibodies" comprised in the present invention are those in which the constant region has been modified or changed from that of the original antibody to generate the properties according to the invention, especially with respect to the binding of C1q and / or to the Fc receptor binding (FcR). This type of chimeric antibody is also called "exchanged class antibody." Chimeric antibodies are the product of expressed immunoglobulin genes that comprise DNA segments encoding immunoglobulin variable regions and DNA segments encoding immunoglobulin constant regions. Methods for producing chimeric antibodies involve conventional recombinant DNA and gene transfection techniques that are well known in the art (see, for example, Morrison SL et al., Proc. Natl. Acad. Sci. USA 81: 6851-6855, 1984; US Patents 5,202,238 and 5,204,244).

The term "humanized antibody" refers to antibodies in which the framework or "complementarity determining regions" (CDR) have been modified to comprise the CDR of an immunoglobulin of specificity different from that of the parental immunoglobulin. In a preferred embodiment, a murine CDR is grafted into the framework region of a human antibody to prepare a "humanized antibody." See, for example, Riechmann, L. et al., Nature 332: 323-327, 1988, and Neuberger, M.S. et al., Nature 314: 268-270, 1985. Particularly preferred CDRs correspond to those representing sequences that recognize the aforementioned antigens of the chimeric antibodies. Other forms of "humanized antibodies" comprised within the present invention are those in which the constant region has been further modified or changed from that of the original antibody to generate the properties according to the invention, especially with respect to the binding of C1q and / or that of the Fc receptor (FcR).

The term "human antibody", as used herein, is intended to include antibodies that have variable and constant regions derived from human germline immunoglobulin sequences. Human antibodies are well known in the state of the art (van Dijk, M.A. and van de Winkel, J.G., Curr. Opin. Chem. Biol. 5: 368-374, 2001). Human antibodies can also be produced in transgenic animals (eg mice) that are capable, after immunization, of producing a complete repertoire or selection of human antibodies in the absence of endogenous immunoglobulin production. The transfer of the human germline immunoglobulin gene series in such germline mutant mice will result in the production of human antibodies after antigen challenge (see, for example, Jakobovits A. et al., Proc. Natl. Acad. Sci. USA 90: 2551-2555, 1993; Jakobovits A. et al., Nature 362: 255-258, 1993; Brueggemann M. et al., Year Immunol. 7 (1993) 33-40). Human antibodies can also be produced in phage expression libraries (Hoogenboom, HR and Winter, G., J. Mol. Biol. 227: 381-388, 1992; Marks, JD et al., J. Mol. Biol. 222: 581 -597, 1991). The techniques of Cole et al. and de Boerner et al. They are also available for the preparation of human monoclonal antibodies (Cole SPC et al., Monoclonal Antibodies and Cancer Therapy, Alan R. Liss, pages 77 to 96, 1985, and Boerner P. et al., J. Immunol. 147: 86-95, 1991). As indicated for the chimeric and humanized antibodies according to the invention, the term "human antibody" as used herein also comprises antibodies modified in the constant region to generate the properties according to the invention, especially with respect to the invention. C1q binding and / or FcR binding, for example by "class exchange", ie, the change or mutation of parts of Fc (for example from IgG1 to IgG4 and / or the IgG1 / IgG4 mutation).

The term "human recombinant antibody", as used herein, is intended to include all human antibodies that are prepared, expressed, created or isolated by recombinant means, such as antibodies isolated from a host cell, such as an NS0 cell. or CHO, or of an animal (eg a mouse) that is transgenic for human immunoglobulin genes or for antibodies expressed using a recombinant expression vector transfected into a host cell. Such human recombinant antibodies have variable and constant regions in a reorganized form. The human recombinant antibodies according to the invention have undergone somatic hypermutation in vivo. Thus, the amino acid sequences of the VH and VL regions of the recombinant antibodies are sequences that, although derived and related to VH and VL sequences of the human germ line, may not naturally exist in the in vivo repertoire of antibodies of the human germ line.

The term "variable domain" (variable domain of a light chain (VL), variable region of a heavy chain (VH)) as used herein refers to each chain of the light and heavy chain pair that participates directly at the binding of the antibody to the antigen. The domains of the human light and heavy variable chains have the same general structure and each domain comprises four framework regions (FR) whose sequences are widely conserved and connected by three "hypervariable regions" (or complementarity determining regions, CDRs). The frame regions adopt a sheet conformation � and the CDRs can form loops that connect the sheet structure �. The CDRs in each chain are maintained

in its three-dimensional structure by the framework regions and together with the CDRs of the other chain form the antigen binding site. The CDR3 regions of the heavy and light antibody chains play a particularly important role in the binding specificity / affinity of the antibodies according to the invention and, therefore, provide a further object of the invention.

The terms "hypervariable region" or "antigen binding part of an antibody", as used herein refer to the amino acid residues of an antibody that are responsible for antigen binding. The hypervariable region comprises amino acid residues of the "complementarity determining regions" or "CDR". The "frame" or "FR" regions are those regions of variable domain different from the residues of the hypervariable region as defined herein. Therefore, the light and heavy chains of an antibody comprise, from N-terminal to C-terminal, domains FR1, CDR1, FR2, CDR2, FR3, CDR3 and FR4. The CDRs in each chain are separated by said framework amino acids. In particular, the heavy chain CDR3 is the region that contributes most to antigen binding. The CDR and FR regions are determined according to the standard definition of Kabat et al., Sequences of Proteins of Immunological Interest, 5th edition, Public Health Service, National Institutes of Health, Bethesda, MD, 1991.

As used herein, the term "binding" or "specific binding" refers to the binding of the antibody to an epitope of the antigen in an in vitro assay, preferably in a plasmon resonance assay (BIAcore, GE-Healthcare, Uppsala, Sweden) with purified wild type antigen. Binding affinity is defined by the terms ka (rate constant for the association of the antibody (or antibody or antigen binding protein) of the antibody / antigen complex), kD (dissociation constant) and KD (kD / ka) . The specific binding or binding refers to a binding affinity (KD) of 10-8 moles / l or less, preferably between 10-9 M and

10-13

 moles / l. Thus, a bispecific antigen-binding protein according to the invention specifically binds to each antigen for which it is specific with a binding affinity (KD) of 10-8 moles / at least, preferably between 10-9 and 10. -13 moles / l.

The binding of the antibody to FcyRIII can be investigated by a BIAcore assay (GE-Healthcare Uppsala, Sweden). Binding affinity is defined by the terms ka (rate constant for antibody / antibody antigen complex association), kD (dissociation constant) and KD (kD / ka).

The term "epitope" includes any polypeptide determinant capable of specifically binding an antibody. In certain embodiments, the epitope determinant includes chemically active surface groups of molecules, such as amino acids, saccharide side chains, phosphoryl or sulfonyl and, in certain embodiments, may have specific three-dimensional structural characteristics, or have specific loading characteristics. An epitope is a region of an antigen that binds to an antibody.

In certain embodiments, it is said that an antibody specifically binds to an antigen in the event that it preferentially recognizes its target antigen in a complex mixture of proteins and / or macromolecules.

In a further embodiment, the bispecific antigen-binding protein according to the invention is characterized in that said full-length antibody is from the human IgG1 subclass, or from the human IgG1 subclass with the L234A and L235A mutations.

In a further embodiment, the bispecific antigen-binding protein according to the invention is characterized in that said full-length antibody is from the human IgG2 subclass.

In a further embodiment, the bispecific antigen-binding protein according to the invention is characterized in that said full-length antibody is from the human IgG3 subclass.

In a further embodiment, the bispecific antigen-binding protein according to the invention is characterized in that said full-length antibody is from the human IgG4 subclass or the human IgG4 subclass with the additional S228P mutation.

Preferably, the bispecific antigen-binding protein according to the invention is characterized in that said full-length antibody is from the human IgG1 subclass or the human IgG4 subclass with the additional S228P mutation.

It has now been found that bispecific antigen-binding proteins according to the invention have improved characteristics, such as biological or pharmacological activity, pharmacokinetic properties or toxicity. They can be used, for example, for the treatment of diseases such as cancer.

The term "constant region" as used in the present applications refers to the sum of the domains of an antibody that are not the variable region. The constant region is not directly involved in the

binding of an antigen, although it shows various effector functions. Depending on the amino acid sequence of the constant region of their heavy chains, the antibodies are divided into the following classes: IgA, IgD, IgE, IgG and IgM, and several of them can be further divided into subclasses, such as IgG1, IgG2, IgG3 and IgG4, IgA1 and IgA2. The constant heavy chain regions that correspond to the different classes of antibodies are called a, 5, E, and m, respectively. The constant light chain (CL) regions that can be found in all five classes of antibody are called K (kappa) and A (lambda).

The term "constant region derived from a human origin" as used in the present application refers to a heavy chain constant region of a human antibody of subclass IgG1, IgG2, IgG3 or IgG4 and / or a constant chain region light kappa or lambda. Such constant regions are well known in the state of the art and are described in, for example, Kabat, E.A. (See, for example, Johnson, G. and Wu, TT, Nucleic Acids Res. 28: 214-218, 2000; Kabat, EA et al., Proc. Natl. Acad. Sci. USA 72: 2785-2788, 1975 .

While the antibodies of the IgG4 subclass show a reduced level of Fc receptor binding (FcyRIIIa), the antibodies of other IgG subclasses show a strong binding. However, Pro238, Asp265, Asp270, Asn297 (loss of Fc carbohydrate), Pro329, Leu234, Leu235, Gly236, Gly237, Ile253, Ser254, Lys288, Thr307, Gln311, Asn434 and His435 are residues that, if altered they also provide a reduced level of Fc binding (Shields, RL et al., J. Biol. Chem. 276: 6591-6604, 2001; Lund, J. et al., FASEB J. 9: 115-119, 1995; Morgan,

A. et al., Immunology 86: 319-324, 1995; EP Patent No. 0 307 434).

In one embodiment, an antigen binding protein according to the invention has a reduced level of FcR binding compared to an IgG1 antibody, and the full length parental antibody is, with respect to FcR binding, of the IgG4 subclass or of subclass IgG1 or IgG2 with a mutation in S228, L234, L235 and / or D265, and / or contains the PVA236 mutation. In one embodiment, the mutations in the full length parental antibody are S228P, L234A, L235A, L235E and / or PVA236. In another embodiment, the mutations in the full length parental antibody are, in IgG4, S228P, and in IgG1, L234A and L235A.

The constant region of an antibody directly participates in the ADCC (antibody-dependent cell-mediated cytotoxicity) and the CDC (complement-dependent cytotoxicity). Complement activation (CDC) is initiated from the binding of complement factor C1q to the constant region of most subclasses of IgG antibody. C1q binding to an antibody is caused by protein-protein interactions defined at the so-called binding site. Such constant region binding sites are known in the state of the art and are described, for example, in Lukas T.J. et al., J. Immunol. 127: 2555-2560, 1981; Brunhouse R. and Zebra J., J. Mol. Immunol 16: 907-917, 1979; Burton D.R. et al., Nature 288: 338-344, 1980; Thommesen J.E. et al., Mol. Immunol 37: 995-1004, 2000; Idusogie E.E. et al., J. Immunol. 164: 4178-4184, 2000; Hezareh M. et al., J. Virol. 75: 1216112168, 2001; Morgan A. et al., Immunology 86: 319-324, 1995, and EP Patent No. 0 307 434. Such constant region binding sites are characterized, for example, by amino acids L234, L235, D270, N297, E318 , K320, K322, P331 and P329 (numbering according to the Kabat EU index).

The term "antibody dependent cell cytotoxicity (ADCC)" refers to the lysis of human target cells by an antigen-binding protein according to the invention in the presence of effector cells. The ADCC is preferably measured by treating an antigen-expressing cell preparation with an antigen-binding protein according to the invention in the presence of effector cells, such as freshly isolated PBMC or purified effector cells from leukocyte layers, for example monocytes or natural killer cells (NK) or an NK cell line of permanent growth.

The term "complement dependent cytotoxicity (CDC)" refers to a process initiated by the binding of complement factor C1q to the Fc part of most IgG antibody subclasses. C1q binding to an antibody is caused by protein-protein interactions defined at the so-called binding site. Said binding sites of the Fc part are known from the prior art (see above). Such binding sites of the Fc part are characterized, for example, by amino acids L234, L235, D270, N297, E318, K320, K322, P331 and P329 (numbering according to the Kabat EU index). The antibodies of the IgG1, IgG2 and IgG3 subclasses usually show complement activation, including C1q and C3 binding, while IgG4 does not activate the complement system and does not bind C1q and / or C3.

The cell-mediated effector functions of the monoclonal antibodies can be enhanced by manipulating their oligosaccharide component, as described in Umana P. et al., Nature Biotechnol. 17: 176180, 1999, and US Patent No. 6,602,684. IgG1 type antibodies, the most commonly used therapeutic antibodies, are glycoproteins that have an N-linked glycosylation site conserved in Asn297 in each CH2 domain. The two biantennial complex oligosaccharides bound to Asn27 are buried between the CH2 domains, forming extensive contacts with the polypeptide skeleton, and their presence is essential for the antibody to mediate effector functions, such as antibody-dependent cellular cytotoxicity (ADCC) ( Lifely MR et al., Glycobiology 5: 813-822, 1995; Jefferis R. et al., Immunol. Rev. 163: 59-76, 1998; Wright

A. and Morrison S.L., Trends Biotechnol. 15 (1997) 26-32). Umana, P. et al. Nature Biotechnol. 17: 176-180, 1999, and WO 99/54342 showed that overexpression in Chinese hamster ovary (CHO) cells of the

(1,4) -N-Acetylglucosaminyltransferase III ("GnTIII"), a glucosyltransferase that catalyzes the formation of bisected oligosaccharides, significantly increased the in vitro ADCC activity of the antibodies. Alterations in the composition of Asn297 carbohydrate or its elimination also affect the binding to FcyR and C1q (Umana P. et al., Nature Biotechnol. 17: 176-180, 1999; Davies J. et al., Biotechnol. Bioeng .74: 288-294, 2001; Mimura Y. et al., J. Biol. Chem. 276: 45539-45547, 2001; Radaev S. et al., J. Biol. Chem. 276: 16478-16483, 2001 ; Shields RL et al., J. Biol. Chem. 276: 6591-6604, 2001; Shields RL et al., J. Biol. Chem. 277: 26733-26740, 2002; Simmons LC et al., J. Immunol Methods 263: 133-147, 2002.

Methods for enhancing cell-mediated effector functions of monoclonal antibodies are reported in, for example, WO No. 2005/018572, No. 2006/116260, No. 2006/114700, No. 2004/065540, No. 2005/011735, No. 2005/027966, No. 1997/028267, US No. 2006/0134709, No. 2005/0054048, No. 2005/0152894, and WO No. 2003/035835 and No. 2000/061739.

In a preferred embodiment of the invention, the bispecific antigen-binding protein is glycosylated (in case it comprises an Fc part of the IgG1, IgG2, IgG3 or IgG4 subclass, preferably of the IgG1 or IgG3 subclass) with a chain saccharide in Asn297, so that the proportion of fucose in said saccharide chain is 65% or less (numbering according to Kabat). In another embodiment, the amount of fucose within said saccharide chain is between 5% and 65%, preferably between 20% and 40%. The term "Asn297" according to the invention refers to the amino acid asparagine located at approximately position 297 in the Fc region. Based on minor variations of the antibody sequence, Asn297 can also be located at some amino acids (usually no more than 63 amino acids) upstream or downstream of position 297, that is, between positions 294 and 300. In one embodiment, The glycosylated antigen binding protein according to the invention is of a subclass of IgG, is of the human IgG1 subclass, of the human IgG1 subclass with the mutations L234A and L235A or of the IgG3 subclass. In a further embodiment, the amount of Nglucolylneuraminic acid (NGNA) is 1% or less, and / or the amount of N-terminal alpha-1,3-galactose is 1% or less within said saccharide chain. The saccharide chain preferably shows the characteristics of the N-linked glycans bound to Asn297 of an antibody recombinantly expressed in a CHO cell.

The expression "saccharide chains show characteristics of N-linked glycans bound to Asn297 of an antibody recombinantly expressed in a CHO cell" refers to the fact that the saccharide chain in Asn297 of the full length parental antibody according to the invention has the same structure and sequence of saccharide residues, except for the fucose residue, than those of the same antibody expressed in unmodified CHO cells, for example such as those reported in WO 2006/103100.

The term "NGNA" as used in the present application refers to the saccharide residue of Nglucolylneuraminic acid.

The glycosylation of human IgG1 or IgG3 occurs in Asn297 in the form of glycosylation of fucosylated biantennial oligosaccharide complex terminated in at most two Gal residues. The human heavy chain constant regions of the IgG1 or IgG3 subclass are reported in detail in Kabat E.A. et al., Sequences of Proteins of Immunological Interest, 5th edition, Public Health Service, National Institutes of Health, Bethesda, MD, 1991, and in Brueggemann M. et al., J. Exp. Med. 166: 1351-1361, 1987; Love T.W. et al., Methods Enzymol. 178 (1989) 515-527. These structures are called G0, G1 (a-1,6- or a-1,3-) or G2 glycan residues, depending on the amount of terminal Gal residues (Raju TS, Bioprocess Int. 1: 44-53, 2003 ). CHO glycosylation of antibody Fc parts is described in, for example, Routier, FH, Glycoconjugate J. 14: 201-207, 1997. Antibodies that are recombinantly expressed in non-glucomodified CHO host cells are usually fucosylated in Asn297 in a proportion of at least 85%. The modified oligosaccharides of the full length parental antibody can be hybrid or complex. Preferably, the bifurcated, reduced / non-fucosylated oligosaccharides are hybrids. In another embodiment, the reduced / non-fucosylated bifurcated oligosaccharides are complex.

According to the invention, the term "amount of fucose" refers to the amount of said sugar in the saccharide chain in Asn297, based on the sum of all glucostructures bound to Asn297 (for example complex, hybrid and high mannose structures ) according to measurement made using MALDI-TOF mass spectrometry and calculated as average value. The relative amount of fucose is the percentage of structures that contain fucose in relation to all the glycostructures identified in a sample treated with N-glucosidase F (for example complex structures, hybrids and structures of low and high mannose content, respectively) according to MALDI-TOF.

The antigen binding protein according to the invention is produced by recombinant means. Thus, one aspect of the present invention is a nucleic acid encoding the antigen binding protein according to the invention, and an additional aspect is a cell comprising said nucleic acid encoding a protein of

antigen binding according to the invention. The methods for recombinant production are widely known in the state of the art and comprise the expression of proteins in prokaryotic and eukaryotic cells with the subsequent isolation of the antigen-binding protein and usually purification to a pharmaceutically acceptable purity. For expression of the antibodies as indicated above in a host cell, nucleic acids encoding the respective modified light and heavy chains are inserted into the expression vectors by standard methods. Expression is carried out in appropriate prokaryotic or eukaryotic host cells, for example CHO cells, NS0 cells, SP2 / 0 cells, HEK293 cells, COS cells, PER.C6 cells, yeast or E. coli cells, and the protein Antigen binding is recovered from the cells (from the supernatant or from the cells after lysis). General methods for recombinant antibody production are well known in the state of the art and are described, for example, in review articles by Makrides, S.C., Protein Expr. Purif. 17: 183-202, 1999; Geisse, S. et al., Protein Expr. Purif. 8: 271-282, 1996; Kaufman, R.J., Mol. Biotechnol 16: 151-160, 2000; Werner, R.G., Drug Res. 48 (1998) 870-880.

The bispecific antigen-binding proteins according to the invention are conveniently separated from the culture medium by conventional immunoglobulin purification procedures such as, for example, A-sepharose protein, hydroxylapatite chromatography, gel electrophoresis, dialysis or affinity chromatography. The DNA or RNA encoding the monoclonal antibodies is isolated and easily sequenced using conventional procedures. Hybridoma cells can serve as a source of said DNA and RNA. After isolation, the DNA can be inserted into expression vectors, which are then transfected into host cells, such as HEK 293 cells, CHO cells or myeloma cells that would not otherwise produce immunoglobulin protein, to obtain the synthesis of recombinant monoclonal antibodies. in the host cells.

Variants (or mutants) of the amino acid sequence of the bispecific antigen-binding protein are prepared by introducing appropriate nucleotide changes into the DNA of the antigen-binding protein, or by nucleotide synthesis. However, these modifications can only be carried out under very limited conditions, for example as indicated above. For example, modifications that do not alter the aforementioned characteristics of the antibody, such as the isotype and antigen binding of IgG but that could improve recombinant production performance, protein stability or facilitate purification.

The term "host cell" as used in the present application refers to any type of cellular system that can be manipulated to generate the antibodies according to the present invention. In one embodiment, HEK293 and CHO cells are used as host cells. As used herein, the terms "cell", "cell line" and "cell culture" are used interchangeably and all such expressions include progeny. Thus, the terms "transformants" and "transformed cells" include the primary subject cell and the cultures derived therefrom regardless of the number of transfers. It should also be interpreted that the entire progeny may not be exactly identical in their DNA content, due to deliberate or natural mutations. It is included the variant progeny that has the same function or biological activity according to the screening performed in the originally transformed cell.

The expression in NS0 cells is described in, for example, Barnes L.M. et al., Cytotechnology 32: 109-123, 2000; Barnes L.M. et al., Biotech. Bioeng 73: 261-270, 2001. Transient expression is described in, for example, Durocher Y. et al., Nucl. Acids Res. 30: E9, 2002. Cloning of variable domains is described in Orlandi, R. et al., Proc. Natl. Acad. Sci. USA 86: 3833-3837, 1989; Carter, P. et al., Proc. Natl. Acad. Sci. USA 89: 4285-4289, 1992, and in Norderhaug, L. et al., J. Immunol. Methods 204: 77-87, 1997. A preferred transient expression system (HEK293) is described in Schlaeger, E.-J. and Christensen, K., Cytotechnology 30: 71-83, 1999, and in Schlaeger, E.-J.,

J. Immunol. Methods 194: 191-199, 1996.

Control sequences that are suitable for prokaryotes include, for example, a promoter, optionally an operator sequence and a ribosomal binding site. It is known that eukaryotic cells use promoters, enhancers and polyadenylation signals.

A nucleic acid is "operably linked" in the event that it is in a functional relationship with another nucleic acid sequence. For example, the DNA of a presequence or secretory leader is operably linked to DNA of a polypeptide if it is expressed as a preprotein that participates in the secretion of the polypeptide; a promoter or enhancer is operably linked to a coding sequence if it affects the transcription of the sequence; or a ribosomal binding site is operably linked to a coding sequence if it is located so as to facilitate translation. Generally, the term "operably linked" refers to the DNA sequences that bind are contiguous and, in the case of a secretory leader, contiguous and in the same reading frame. However, intensifiers are not necessarily contiguous. The union is carried out by site ligation

of appropriate restriction. In the event that such sites do not exist, synthetic oligonucleotide or linker adapters are used according to conventional practice.

Purification of the antibodies is carried out in order to remove cellular components or other contaminants, for example other nucleic acids or cellular proteins, by standard techniques, including alkaline / SDS treatment, band formation in CsCl, column chromatography, agarose gel electrophoresis, and other techniques well known in the art. See Ausubel, F. et al., Editor, Current Protocols in Molecular Biology, Greene Publishing and Wiley Interscience, New York, 1987. A number of different methods are well established and widely used for protein purification, such as affinity chromatography with microbial proteins (for example protein A or protein G affinity chromatography), ion exchange chromatography (for example cation exchange (carboxymethyl resins), anion exchange (aminoethyl resins) and mode exchange mixed), thiophilic adsorption (for example with beta-mercaptoethanol and with other SH ligands), hydrophobic interaction or aromatic adsorption chromatography (for example with phenyl sepharose, aza-arenophilic resins or m-aminophenylboronic acid), metal chelate affinity chromatography (for example with Ni (II) and Cu (II) affinity material), exclusion chromatography by size, and electrophoretic methods (such as gel electrophoresis and capillary electrophoresis) (Vijayalakshmi, M.A., Appl. Biochem Biotech 75: 93-102, 1998).

An aspect of the invention is a pharmaceutical composition comprising an antigen binding protein according to the invention. Another aspect of the invention is the use of an antigen binding protein according to the invention for the preparation of a pharmaceutical composition. A further aspect of the invention is a method for the preparation of a pharmaceutical composition comprising an antigen binding protein according to the invention. In another aspect, the present invention provides a composition, for example a pharmaceutical composition, containing an antigen binding protein according to the present invention, formulated in conjunction with a pharmaceutical carrier.

Another aspect of the invention is said pharmaceutical composition for the treatment of cancer.

Another aspect of the invention is the bispecific antigen-binding protein according to the invention for the treatment of cancer.

Another aspect of the invention is the use of an antigen binding protein according to the invention for the preparation of a medicament for the treatment of cancer. Another aspect of the invention is a method of treating a patient suffering from cancer, by administering an antigen binding protein according to the invention in said patient in need of said treatment.

As used herein, the term "pharmaceutical carrier" includes each and every solvent, dispersion media, coatings, antibacterial and antifungal agents, isotonic and delayed absorption agents and the like that are physiologically compatible. Preferably, the carrier is suitable for intravenous, intramuscular, subcutaneous, parenteral, spinal or epidermal administration (for example by injection or infusion).

A composition of the present invention can be administered by a variety of methods known in the art. As will be appreciated by the person skilled in the art, the route and / or the mode of administration will vary depending on the desired results. To administer a compound of the invention by certain routes of administration, it may be necessary to coat the compound with a material, or co-administer the compound with a material, to prevent its inactivation. For example, the compound can be administered in a subject in an appropriate carrier, for example liposomes or a diluent. Pharmaceutically acceptable diluents include saline solution and aqueous buffer solutions. Pharmaceutical carriers include sterile aqueous solutions or dispersions and sterile powders for the extemporaneous preparation of sterile injectable solutions or dispersions. The use of said media and agents for pharmaceutically active substances is known in the art.

The terms "parenteral administration" and "parenterally administered" as used herein refer to modes of administration other than enteric and topical administration, usually by injection, and include, but are not limited to, injections and infusions. intravenous, intramuscular, intraarterial, intrathecal, intracapsular, intraorbital, intracardiac, intradermal, intraperitoneal, transtracheal, subcutaneous, subcuticular, intraarticular, subcapsular, subarachnoid, intraspinal, epidural and intrasternal

The term "cancer" as used herein refers to proliferative diseases, such as lymphomas, lymphocytic leukemias, lung cancer, non-small cell lung cancer (NSCL), bronchioloalveolar cell lung cancer, bone cancer, cancer. pancreatic, skin cancer, head or neck cancer, cutaneous or intraocular melanoma, uterine cancer, ovarian cancer, rectal cancer, anal region cancer,

stomach cancer, gastric cancer, colon cancer, breast cancer, uterine cancer, carcinoma of the fallopian tubes, endometrial carcinoma, carcinoma of the cervix, carcinoma of the vagina, carcinoma of the vulva, Hodgkin's disease, esophageal cancer , small bowel cancer, endocrine system cancer, thyroid cancer, parathyroid gland cancer, adrenal gland cancer, soft tissue sarcoma, urethral cancer, penile cancer, prostate cancer, bladder cancer, kidney or ureter cancer, renal cell carcinoma, renal pelvis carcinoma, mesothelioma, hepatocellular cancer, biliary cancer, central nervous system (CNS) neoplasms, spinal axis tumors, brain stem glioma, glioblastoma multiforme, astrocytomas, schwanomas , ependinomas, medulloblastomas, meningiomas, squamous cell carcinomas, pituitary adenoma and Ewings sarcoma, including refractory versions of any a of the cancers indicated above, or a combination of one or more of the cancers indicated above.

Such compositions may also contain adjuvants, such as preservatives, wetting agents, emulsifying agents and dispersing agents. The prevention of the presence of microorganisms can be guaranteed both by sterilization procedures, supra, and by the inclusion of various antibacterial and antifungal agents, for example paraben, chlorobutanol, phenol, sorbic acid and the like. It may also be desirable to include isotonic agents, such as sugars, sodium chloride and the like in the compositions. In addition, prolonged absorption of the injectable pharmaceutical form can be achieved by including agents that delay absorption, such as aluminum monostearate and gelatin.

Regardless of the route of administration selected, the compounds of the present invention, which can be used in a suitable hydrated form, and / or the pharmaceutical compositions of the present invention, are formulated in pharmaceutically acceptable dosage forms by conventional methods known by the skilled.

The actual dose levels of the active ingredients in the pharmaceutical compositions of the present invention can be modified so that an amount of the active ingredient is obtained that is effective to achieve the desired therapeutic response in a particular patient, composition and mode of administration, without being toxic to the patient. The dose level selected will depend on a variety of pharmacokinetic factors, including the activity of the particular compositions of the present invention used, the route of administration, the time of administration, the rate of excretion of the particular compound used, the duration of treatment, other drugs, the compounds and / or materials used in combination with the particular compositions used, age, sex, weight, condition, general health status and medical history of the patient under treatment, and similar factors well known in the art medical

The composition must be sterile and liquid enough to make the composition administrable by syringe. In addition to water, the carrier is preferably an isotonic buffered saline solution.

The correct fluidity can be maintained, for example, by the use of a coating such as lecithin, by the maintenance of a required particle size in the case of a dispersion, and by the use of surfactants. In many cases, it is preferable to include isotonic agents, for example sugars, polyalcohols, such as mannitol or sorbitol, and sodium chloride in the composition.

As used herein, the terms "cell", "cell line" and "cell culture" are used interchangeably and all such expressions include progeny. Thus, the terms "transformants" and "transformed cells" include the primary subject cell and the cultures derived therefrom regardless of the number of transfers. It should also be interpreted that the entire progeny may not be exactly identical in their DNA content, due to deliberate or natural mutations. It is included the variant progeny that has the same function or biological activity according to the screening performed in the originally transformed cell. Where different denominations are intended to be used, these will be evident from the context.

The term "transformation" as used herein refers to a method of transferring a vector / nucleic acid into a host cell. In the case that cells without strong cell wall barriers are used as host cells, transfection is carried out, for example, by the calcium phosphate precipitation method, as described by Graham F.L. and Van der Eb A.J., Virology 52: 456-467, 1973. However, other methods can also be used to introduce DNA into cells, such as nuclear injection or by fusion of protoplasts. In the case where prokaryotic cells or cells containing substantial cell wall constructs are used, a method of transfection is, for example, treatment with calcium using calcium chloride as described in Cohen S.N. et al., PNAS 69 (1972) 2110-2114.

As used herein, the term "expression" refers to the procedure by which a nucleic acid is transcribed into mRNA and / or to the procedure by which the transcribed mRNA (also called transcribed) is subsequently translated into peptides, polypeptides or proteins. Transcripts and polypeptides

coded are collectively called "gene product". In the event that the genomic DNA polynucleotide is derived, expression in a eukaryotic cell may include mRNA processing. A "vector" is a nucleic acid molecule, particularly self-replicative, that transfers a nucleic acid molecule inserted into or inside host cells. The term includes vectors that work primarily by inserting DNA or RNA into a cell (for example chromosomal integration), the replication of vectors that work primarily by replicating DNA or RNA, and expression vectors that work by transcribing and / or translating DNA or RNA. Also included are vectors that provide more than one of the functions described above.

An "expression vector" is a polynucleotide that, when introduced into an appropriate host cell, can be transcribed and translated into a polypeptide. An "expression system" usually refers to a suitable host cell comprising an expression vector that can function by yielding a desired expression product.

The following examples, sequence listing and figures are provided to aid in the understanding of the present invention, the actual scope of which is provided in the appended claims.

Description of the sequence listing

SEQ ID No. 1 <Ang-2> of unmodified heavy chain with fused VH-CL domains of <VEGF> C

terminal of modified Fab fragment (CH1-CL exchange) SEQ ID No. 2 domains VL-CH1 of <VEGF> of modified Fab fragment (CH1-CL exchange) SEQ ID No. 3 <Ang-2> of unmodified light chain SEQ ID No. 4 <Ang-2> heavy chain unmodified with merged VH-CL domains of <VEGF> N

terminal of modified Fab fragment (CH1-CL exchange) SEQ ID No. 5 <VEGF> of modified heavy chain (CH1-CL exchange) with VH-CH1 domains of <Ang-2> fused C-terminally of unmodified Fab fragment

Description of the figures

Figure 1 Schematic structure of a full-length antibody without specific binding of the CH4 domain to a first antigen 1 with two heavy chain and light chain partners comprising variable and constant domains in a typical order.

Figures 2a and 2b Schematic structure of typical unmodified Fab fragments that specifically bind a second antigen 2 with the linker peptide at the Cterminal end (Figure 2a) or at the N-terminal end (Figure 2b) of the CH1-VH chain .

Figure 3 Schematic structure of a full-length antibody that specifically binds to a first antigen 1 to which two unmodified Fab fragments that specifically bind to a second antigen 2 have been fused at the N-terminus of their heavy chain (Fig. 3a ) and unwanted by-products due to incorrect mating (figs. 3b and fig. 3c).

Figures 4a, 4b and 4c Schematic structure of bispecific antigen-binding proteins according to the invention in which the incorrect pairing has been reduced by replacing certain domains a) in the Fab fragment of the full-length antibody that specifically binds to a first antigen 1 and / or b) in the two additional fused Fab fragments that specifically bind to a second antigen 2. Figure 4a shows bispecific antigen-binding proteins. Figures 4b and 4c show all combinations of VH / VL and / or CH1 / CL domain exchanges within the full length Fab fragments and the full length Fab fragments leading to bispecific antigen-binding proteins according to the invention which they have a reduced level of incorrect mating.

Figure 5 Schematic structure of bispecific antigen-binding proteins according to the invention that recognize Ang-2 and VEGF (Example 1).

Figure 6 Schematic structure of bispecific antigen-binding proteins according to the invention that recognize Ang-2 and VEGF (Example 2).

Figure 7 Schematic structure of bispecific antigen-binding proteins according to the invention that recognize Ang-2 and VEGF (Example 3).

Examples

General materials and methods

5 General information on nucleotide sequences of light and heavy chains of human immunoglobulins is provided in: Kabat E.A. et al., Sequences of Proteins of Immunological Interest, 5th edition, Public Health Service, National Institutes of Health, Bethesda, MD, 1991. The amino acids of the antibody chains are numbered and referenced according to EU numbering ( Edelman GM et al., Proc. Natl. Acad. Sci. USA 63: 78-85, 1969; Kabat EA et al., Sequences of Proteins of Immunological Interest, 5th edition, Public Health

10 Service, National Institutes of Health, Bethesda, MD, 1991).

Recombinant DNA Techniques

Standard methods are used to manipulate DNA, as described in Sambrook J. et al., Molecular

15 cloning: A laboratory manual; Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York, 1989. Molecular biological reagents were used according to the manufacturer's instructions.

Gene synthesis

20 Desired gene segments were prepared from oligonucleotides constructed by chemical synthesis. The gene segments, which were flanked by single endonuclease cleavage sites, were assembled by hybridization and oligonucleotide ligation, including PCR amplification, and subsequently cloned by the indicated restriction sites, for example KpnI / SacI or AscI / PacI in a pGA4 cloning vector based on pPCRScript (Stratagene). The DNA sequences of the subcloned gene fragments are

25 confirm by DNA sequencing. The gene synthesis fragments are ordered according to the specifications provided by Geneart (Regensburg, Germany).

DNA sequence determination

30 DNA sequences were determined by double chain sequencing performed at MediGenomix GmbH (Martinsried, Germany) or at Sequiserve GmbH (Vaterstetten, Germany).

DNA and protein sequence analysis and sequence data management

35 The GCG software package (Genetics Computer Group, Madison, Wisconsin), version 10.2, and the Infomax Vector NT1 Advance suite version 8.0 software were used for sequence creation, mapping, analysis, annotation and illustration.

Expression vectors

For the expression of the indicated tetravalent bispecific antibodies, expression plasmid variants were used for transient expression cells (for example in HEK293 EBNA or HEK293-F cells) based on a cDNA organization with or without a CMV intron A promoter or in a genomic organization with a CMV promoter.

Apart from the antibody expression cassette, the vectors contained:

-

an origin of replication that allows replication of said plasmid in E. coli, and

-

a -lactamase gene that provides ampicillin resistance to E. coli.

The antibody gene transcription unit is composed of the following elements:

-

one or more unique restriction sites at the 5 ’end

-

the immediate early enhancer and promoter of human cytomegalovirus, followed by the intron A sequence in the case of DNA organization,

-

a 5 ’untranslated region of a human antibody gene,

-

an immunoglobulin heavy chain signal sequence,

-

the human tetravalent bispecific antibody chain (wild type or domain exchange)

in the form of cDNA or genomic organization with the exon-intron organization of immunoglobulin 60 - a 3 ’region not translated with a polyadenylation signal sequence, and

-

one or more unique restriction sites at the 3 ’end.

Fusion genes comprising the indicated antibody chains, as indicated below, are generated by PCR and / or gene synthesis, and assembled using known recombinant methods and techniques by connecting the corresponding nucleic acid segments, for example using unique restriction sites in the respective vectors. Subcloned nucleic acid sequences were verified by DNA sequencing. For transient transfections, larger amounts of the plasmids were prepared from transformed E. coli cultures (Nucleobond AX, Macherey-Nagel).

Cell culture techniques

Standard cell culture techniques such as those described in Current Protocols in Cell Biology, Bonifacino J.S., Dasso M., Harford J.B., Lippincott-Schwartz J. and Yamada K.M. were used. (editors), John Wiley & Sonds, Inc., 2000.

Tetravalent bispecific antibodies were expressed by transient cotransfection of the three respective expression plasmids in HEK293-EBNA cells in adherent growth or in HEK29-F cells growing in suspension, as described below.

Transient transfections in the HEK293-EBNA system

Tetravalent bispecific antibodies were expressed by transient cotransfection of the three respective expression plasmids (for example encoders of the modified heavy chain, as well as the corresponding light and modified modified chains) in adherent growth HEK293-EBNA cells (human embryonic renal 293 cell line expressing Epstein-Barr virus nuclear antigen, deposit number of the American Type Culture Collection (ATCC) No. CRL-10852, lot 959 218) grown in DMEM (Eagle medium modified by Dulbecco, Gibco) supplemented with FCS (serum 10% bovine fetus, Gibco) with ultra-low levels of IgG, 2 mM L-glutamine (Gibco) and 250 mg / ml geneticine (Gibco). For transfection, FuGENETM 6 (Roche Molecular Biochemicals) transfection reagent was used in a ratio of FuGENETM reagent (μl) to DNA (μg) of 4: 1 (between 3: 1 and 6: 1). Proteins were expressed from the respective plasmids using a molar ratio of light chain (modified and wild-type) and heavy chain coding plasmids of 1: 1: 1 (equimolar) between 1: 1: 2 and 2: 2: 1 respectively. The cells are fed on day 3 with 4 mM L-glutamine, glucose [Sigma] and NAA [Gibco]. Cell culture supernatants containing tetravalent bispecific antibody were collected between days 5 and 11 after transfection by centrifugation and stored at -20 ° C. General information on recombinant expression of human immunoglobulins is provided in, for example, HEK293 cells, in: Meissner P. et al., Biotechnol. Bioeng 75: 197-203, 2001.

Transient transfections in the HEK293-F system

Alternatively, tetravalent bispecific antibodies were generated by transient transfection of the respective plasmids (for example encoders of the modified heavy chain, as well as the corresponding modified light and light chains) using the HEK293F system (Invitrogen) following the manufacturer's instructions. Briefly, HEK293-F (Invitrogen) cells cultured in suspension in a shake flask or in a fermenter under agitation were transfected into serum-free FreeStyle 293 expression medium (Invitrogen) transfected with a mixture of the three expression plasmids such as indicated above and effect

or fectin (Invitrogen). For a 2-liter oscillating flask (Corning), HEK293-F cells were seeded at a density of 106 cells / ml in 600 ml and incubated at 120 rpm, in 8% CO2. The days after transfecting the cells at a cell density of approximately 1.5x106 cells / ml with approximately 42 ml of A mixture) 20 ml of Opti-MEM (Invitrogen) with 600 μg of total plasmid DNA (1 μg / ml ) modified heavy chain coding, corresponding light chain and corresponding modified light chain, in equimolar proportion, and B) 20 ml of Opti-MEM + 1.2 ml of 293fectin or fectin (2 μl / ml). Depending on glucose consumption, glucose solution was added during the course of fermentation. After 5 to 10 days the supernatant containing the secreted antibody was collected, and the antibodies were purified directly from the supernatant or the supernatant was frozen and stored.

Protein determination

The concentration of purified tetravalent bispecific antibody proteins and derivatives thereof was determined by determining the optical density (OD) at 280 nm, using the molar extinction coefficient calculated based on the amino acid sequence, according to Pace C.N. et al., Protein Science 4: 2411-2423, 1995.

Determination of antibody concentration in supernatants

The concentration of tetravalent bispecific antibodies in cell culture supernatants was estimated by immunoprecipitation with A-agarose protein beads (Roche). 60 µl of agarose-protein A beads were washed

three times in TBS-NP40 (50 mM Tris, pH 7.5, 150 mM NaCl, 1% Nonidet-P40). Subsequently, 1 to 15 ml of cell culture supernatant was applied to the pre-equilibrated agarose-protein A beads in TBS-NP40. After incubation for 1 hour at room temperature, the beads were washed on an Ultrafree-MC filter column (Amicon) once with 0.5 ml of TBS-NP40, twice with 0.5 ml of 2x buffered saline with phosphate (2xPBS, Roche) and briefly four times with 0.5 ml of 100 mM sodium citrate, pH 5.0. The ligated antibody was eluted by the addition of 35 µl buffer for NuPAGE® LDS samples (Invitrogen). Half of the sample was combined with NuPAGE® sample reducing agent or left unchecked, respectively, and heated for 10 minutes at 70 ° C. Then, 5 to 30 μl were applied to a 4-12% Bis-Tris NuPAGE® BisS-PAGE (Invitrogen) (with MOPS buffer for non-reduced SDS-PAGE and MES buffer with NuPAGE® migration buffer antioxidant additive (Invitrogen) ) for a reduced SDS-PAGE) and stained with Coomassie blue.

The concentration of tetravalent bispecific antibodies in cell culture supernatants was measured quantitatively by HPLC affinity chromatography. Briefly, cell culture supernatants containing antibodies and derivatives that bind protein A were applied to a Poros A / 20 column of Applied Biosystems in 200 mM KH2PO4, 100 mM sodium citrate, pH 7.4, and eluted from the 200 mM NaCl matrix, 100 mM citric acid, pH 2.5 in an Agilent 1100 HPLC system. Eluted proteins were quantified by UV absorbance and integration of peak areas. A purified standard IgG1 antibody served as standard.

Alternatively, the concentration of tetravalent bispecific antibodies in cell culture supernatants was measured by IgG sandwich ELISA. Briefly, 96-well streptavidin StreptaWell High Bind (Roche) coated microtiter plates were coated with 100 µl / well of biotinylated anti-human IgG capture molecule F (ab ') 2 <h-Fcy> BI (Dianova) at concentration of 0.1 μg / ml for 1 hour at room temperature, or alternatively overnight at 4 ° C and subsequently washed three times with 200 μl / well of PBS, 0.05% Tween (PBST, Sigma). 100 µl / well of a dilution series in PBS (Sigma) of the respective cell culture supernatants containing the antibody were added to the wells and incubated for 1 to 2 hours on a microtiter plate stirrer at room temperature. The wells were washed three times with 200 μl / well of PBST and the bound antibody was detected with 100 μl of F (ab ') 2 <hFcy> POD (Dianova) at a concentration of 0.1 μg / ml as detection antibody for 1 to 2 hours on an agitator for microtiter plates at room temperature. Unbound detection antibody was washed three times with 200 µl / well of PBST and bound detection antibody was detected by adding 100 µl of ABTS / well. The absorbance was determined on a Tecan Fluor spectrometer at a measurement wavelength of 405 nm (reference wavelength: 492 nm).

Protein purification

Proteins were purified from filtered cell culture supernatants following standard protocols. Briefly, tetravalent bispecific antibodies were applied to an A-sepharose protein (GE Healthcare) column and washed with PBS. Elution of tetravalent bispecific antibodies was achieved at pH 2.8, followed by immediate neutralization of the sample. Aggregate proteins were separated from monomeric antibodies by size exclusion chromatography (Superdex 200, GE Healthcare) in PBS or 20 mM histidine, 150 mM NaCl, pH 6.0. The monomeric fractions were pooled, concentrated if necessary using, for example, an Amicon Ultra centrifuge concentrator (cut-off value: 30 MWCO) from Millipore, frozen and stored at -20 ° C or -80 ° C. Part of the samples were reserved for subsequent protein analysis and for analytical characterization, for example by SDS-PAGE, size exclusion chromatography or mass spectrometry.

SDS-PAGE

The NuPAGE® pre-molded gel system (Invitrogen) was used following the manufacturer's instructions. In particular, NuPAGE® Novex® Bis-TRIS Pre-Cast gels (pH 6.4) and a NuPage® MES run buffer (reduced gels with NuPAGE® antioxidant additive for run buffer) or MOPS (non-reduced gels) were used .

Analytical chromatography by size exclusion

Size exclusion chromatography for the determination of aggregation and oligomeric status of tetravalent bispecific antibodies was carried out by HPLC chromatography. Briefly, protein A purified antibodies were applied on a Tosoh TSKgel G3000SW column in 300 mM NaCl, 50 mM KH2PO4 / K2HPO4, pH 7.5 in an Agilent 1100 HPLC system or in a Superdex 200 column (GE Healthcare) in 2xPBS in a Dionex HPLC system. Eluted proteins were quantified by UV absorbance and integration of peak areas. BioRad 151-1901 gel filtration standard served as a standard.

Mass spectrometry

The total deglycosylated mass of tetravalent bispecific antibodies was determined and confirmed by electrospray ionization mass spectrometry (EM-IEP). Briefly, 100 mg of antibodies purified with 50 mM N-glucosidase F (PNGasaF, ProZyme) in 100 mM KH2PO4 / K2HPO4, pH 7, at 37 ° C for 12 to 24 hours at a protein concentration of up to 2 mg / ml and subsequently they were desalted by HPLC on a Sephadex G25 column (GE Healthcare). The mass of the respective modified, light and light modified heavy chains was determined by ESI-MS after deglycosylation and reduction. Briefly, 50 Ig of tetravalent bispecific antibody was incubated in 115 μl, with 60 μl of 1 M TCEP and 50 μl of 8 M guanidine hydrochloride subsequently desalted. The total mass and mass of the heavy and light chains reduced was determined by ESI-MS in a Q-Star Elite MS system equipped with a NanoMate source.

VEGF binding ELISA

The binding properties of tetravalent bispecific antibodies were evaluated in an ELISA assay with full length VEGF165-His protein (R&D Systems). For this, Falcon enhanced transparent polystyrene microtiter plates were coated with 100 µl of 2 µg / ml recombinant human VEGF165 (R&D Systems) in PBS for 2 hours at room temperature or overnight at 4 ° C. The wells were washed three times with 300 μl of PBST (0.2% Tween-20) and blocked with 200 μl of 2% BSA-0.1% Tween-20 for 30 minutes at room temperature and subsequently washed three times with 300 μl of PBST. 100 μl / well of a dilution series of tetravalent bispecific antibodies purified in PBS (Sigma) were added to the wells and incubated for 1 hour in a microtiter plate stirrer at room temperature. Wells were washed three times with 300 μl of PBST (0.2% Tween-20) and bound antibody was detected with 100 μl / well of F (ab ') 0.1 mg / ml <Fc-gamma_h> POD (Immunoresearch) in 2% BSA-0.1% Tween-20 as a detection antibody for 1 hour on a microtiter plate stirrer at room temperature. Unbound detection antibody was washed three times with 300 µl / well of PBST and bound detection antibody was detected by the addition of 100 µl of ABTS / well. The absorbance was determined on a Tecan Fluor spectrometer at a measurement wavelength of 405 nm (reference wavelength: 492 nm).

VEGF binding: Kinetic characterization of VEGF binding at 37 ° C by surface plasmon resonance (Biacore)

In order to further corroborate the ELISA results, the binding of tetravalent bispecific antibodies to VEGF was quantitatively analyzed using surface plasmon resonance technology in a Biacore T100 instrument according to the following protocol and using the T100 software package. Briefly, tetravalent bispecific antibodies were captured on a CM5 chip by binding to goat anti-human IgG (JIR 109-005-098). The capture antibody was immobilized by coupling of amino groups using standard coupling of aminos, as follows: The HBS-N buffer served as a run buffer, activation was carried out with the EDC / NHS mixture with the aim of a 700 RU ligand density. The capture antibody was diluted in NaAc coupling buffer, pH 5.0, c = 2 μg / ml, finally the activated carboxyl groups were blocked by injection of 1 M ethanolamine. The capture of tetravalent bispecific <VEGF> antibodies it was carried out at a flow rate of 5 μl / minute and c = 10 nM, diluted with migration buffer + 1 mg / ml BSA; a catch level of approximately 30 RU should be achieved. VEGFrh (VEGFrh, R&D Systems Cat. No. 293-VE) was used as an analyte. Kinetic characterization of VEGF binding to tetravalent bispecific <VEGF> antibodies was carried out at 25 ° C or 37 ° C in PBS + 0.005% Tween-20 (v / v) as run buffer. The sample was injected at a flow rate of 50 μl / minute and an association time of 80 seconds and a dissociation time of 1,200 seconds, with a series of VEGFrh concentrations of 300 to 0.29 nM. The regeneration of the free capture antibody surface was carried out with 10 mM glycine, pH 1.5, and a contact time of 60 seconds after each analyte cycle. The kinetic constants were calculated using the usual double reference method (control reference: binding of VEGFrh to human anti-IgG goat antibody capture molecule, targets in the measurement flow cell, VEGFrh concentration "0" , model: Langmuir union 1: 1 (Rmax was set to local due to the binding of capture molecules).

Ang-2 binding ELISA

The binding properties of tetravalent bispecific antibodies were evaluated in an ELISA assay with full-length Ang-2-His protein (R&D Systems). To do this, Falcon enhanced transparent polystyrene microtiter plates were coated with 100 µl of 1 µg / ml recombinant human Ang-2 (R&D Systems, no carrier) in PBS for 2 hours at room temperature or overnight at 4 ° C. The wells were washed three times with 300 μl of PBST (0.2% Tween-20) and blocked with 200 μl of 2% BSA-0.1% Tween-20 for 30 minutes at room temperature and subsequently washed three times with 300 μl of PBST. 100 μl / well of a dilution series of tetravalent bispecific antibodies purified in PBS (Sigma) were added to the wells and incubated for 1 hour in a microtiter plate stirrer at room temperature. The wells were washed three times with 300 μl of PBST (0.2% Tween-20) and the bound antibody was detected with 100 μl / well of F (ab ') of <hk> POD (Biozol Cat. No. 206005 ) 0.1 mg / ml in 2% BSA-0.1% Tween-20 as

detection antibody, for 1 hour on a microtiter plate shaker at room temperature. Unbound detection antibody was washed three times with 300 µl / well of PBST and bound detection antibody was detected by the addition of 100 µl of ABTS / well. The absorbance was determined on a Tecan Fluor spectrometer at a measurement wavelength of 405 nm (reference wavelength: 492 nm).

BIACORE binding Ang-2

The binding of tetravalent bispecific antibodies to human Ang-2 was investigated by surface plasmon resonance using a BIACORE T100 instrument (GE Healthcare Biosciences AB, Uppsala, Sweden). Briefly, for affinity measurements, polyclonal goat antibodies <hIgG-Fcy> were immobilized on a CM5 chip via amine coupling for the presentation of tetravalent bispecific antibodies against human Ang-2. Binding was measured in HBS buffer (HBS-P (10 mM HEPES, 150 mM NaCl, 0.005% Tween-20, pH 7.4), 25 ° C. Purified Ang-2-His (R&S Systems or purified on the own laboratory) in various concentrations in solution The association was measured by a 3-minute Ang-2 injection; the dissociation was measured by washing the surface of the chip with HBS buffer for 3 minutes and a KD value was estimated using a 1: 1 Lagmuir binding model Due to the heterogeneity of the Ang-2 preparation, 1: 1 binding could not be observed; thus, KD values are only relative estimates. Negative control data were subtracted ( for example, the buffer curves) of the sample curves for the correction of the intrinsic drift of the system baseline and for the reduction of the noise signal The Biacore T100 Evaluation version 1.1.1 program was used for the analysis of the sensorgrams and for the calculation of the affinity data Alternatively, Ang-2 could be captured at a capture level between 2,000 and 1,700 UR by means of a pentais antibody (PentaHis-Ab without BSA, Qiagen No. 34660) immobilized on a CM5 chip by means of amine coupling (without BSA) (see later).

Ang-2-VEGF Bridge Formation ELISA

The binding properties of tetravalent bispecific antibodies were evaluated in an ELISA assay with immobilized full length VEGF165-His protein (R&D Systems) and human Ang-2-His protein (R&D Systems) for detection of bound bispecific antibody. Only a tetravalent bispecific antibody of <VEGF-Ang-2> was able to bind simultaneously to VEGF and Ang-2, and thus bridge the two antigens, while the "standard" monospecific IgG1 antibodies were not capable of join VEGF and Ang-2 simultaneously (figure 7).

For this, Falcon enhanced transparent polystyrene microtiter plates were coated with 100 µl of 2 µg / ml recombinant human VEGF165 (R&D Systems) in PBS for 2 hours at room temperature or overnight at 4 ° C. The wells were washed three times with 300 μl of PBST (0.2% Tween-20) and blocked with 200 μl of 2% BSA-0.1% Tween-20 for 30 minutes at room temperature and subsequently washed three times with 300 μl of PBST. 100 µl / well of a dilution series of tetravalent bispecific antibodies purified in PBS (Sigma) were added to the wells and incubated for 1 hour in a microtiter plate stirrer at room temperature. The wells were washed three times with 300 µl of PBST (0.2% Tween-20) and bound antibodies were detected by the addition of 100 µl of 0.5 Ang / ml human Ang-2-His (R&D Systems) in PBS. The wells were washed three times with 300 μl of PBST (0.2% Tween-20) and bound Ang-2 was detected with 100 μl of <Ang-2> mIgG1-biotin 0.5 μg / ml antibody ( BAM0981, R&D Systems) for 1 hour at room temperature. Unbound detection antibody was washed three times with 300 µl of PBST (0.2% Tween-20) and bound antibodies were detected by adding 100 µl of streptavidin-POD 1: 2,000 conjugate (Roche Diagnostics GmbH , Cat. No. 11089153) diluted 1: 4 in blocking buffer for 1 h at room temperature. The unbound streptavidin-POD conjugate was washed three-six times with 300 µl of PBST (0.2% Tween-20) and the bound streptavidin-POD conjugate was detected by the addition of 100 µl of ABTS / well. The absorbance was determined on a Tecan Fluor spectrometer at a measurement wavelength of 405 nm (reference wavelength: 492 nm).

Demonstration of simultaneous binding of tetravalent bispecific antibody <VEGF-Ang-2> to VEGF-A and Ang-2 by Biacore

In order to further corroborate the ELISA data of bridging, an additional assay was established to confirm simultaneous binding to VEGF and Ang-2 using surface plasmon resonance technology in a Biacore T100 instrument following the following protocol, and performing the analysis with the T100 software package (T100 Control, version 2.01, T100 Evaluation, version 2.01, T100 Kinetics Summary, version 1.01): Ang-2 was captured with a capture level between 2,000 and 1,700 RU in PBS, 0.005% Tween-20 run buffer (v / v), using a PentaHis antibody (PentaHis-Ab without BSA, Qiagen No. 34660) immobilized on a CM5 chip by coupling amine groups (without BSA). The HBS-N buffer served as a run buffer during the coupling; activation was carried out with a mixture of EDC / NHS. The PentaHis-Ab capture antibody without BSA was diluted in NaAc coupling buffer, pH 4.5, c = 30 μg / ml, finally the carboxyl groups still

activated were blocked by injection of 1 M ethanolamine; ligand densities of between 5,000 and 17,000 UR were tested. Ang-2 was captured at a concentration of 500 nM with the PentaHis-Ab antibody at a flow rate of 5 μl / minute diluted with migration buffer + 1 mg / ml BSA. Next, Ang-2 and VEGF binding of the tetravalent bispecific antibody <Ang-2, VEGF> was demonstrated by incubation with VEGFrh and the formation of a sandwich complex. To this end, the tetravalent bispecific antibody <VEGF-Ang-2> was attached to Ang-2 at a flow rate of 50 μl / minute and at a concentration of 100 nM, diluted with migration buffer + 1 mg / ml BSA and detected simultaneous binding by incubation with VEGF (VEGFrh, R & D-Systems, cat. no. 293-VE) in PBS + 0.005% Tween-20 migration buffer (v / v) at a flow rate of 50 μl / minute and a concentration of 150 nM VEGF. Association time: 120 s; dissociation time: 1,200 s. Regeneration was carried out after each cycle at a flow rate of 50 μl / minute with 2x10 mM glycine, pH 2.0, and a contact time of 60 s. The sensorgrams were corrected using the usual double reference method (control reference: binding of bispecific antibody and VEGFrh to the capture molecule PentaHisAb). The targets were measured for each Ab with VEGFrh concentration "0".

HEK293-Tie2 cell line generation

In order to determine the interference of tetravalent bispecific antibodies <Ang-2, VEGF> with the phosphorylation of Tie2 stimulated by Ang-2 and the binding of Ang-2 to Tie2 on the cells, a recombinant HEK293- cell line was generated Tie. Briefly, a pcDNA3-based plasmid (RB22-pcDNA3 Mole hTie2) encoding full-length human Tie2 was transfected under the control of a CMV promoter and a neomycin resistance marker, using Fugene (Roche Applied Science) as a reagent transfection into HEK293 cells (ATCC) and resistant cells were selected in DMEM + 10% FCS, G418 500 μg / ml. Individual clones were isolated by a cloning ring, and subsequently analyzed for Tie2 expression by FACS. Clone 22 was identified as a clone with high stable expression of Tie2, even in the absence of G418 (HEK293-Tie2 clone 22). Next, HEK293-Tie2 clone 22 was used for the following cell assays: Ang2-induced Tie2 phosphorylation assay and Ang-2 cell ligand binding assay.

Tie2 phosphorylation assay induced by Ang-2

The inhibition of Ang2-induced Tie2 phosphorylation by tetravalent bispecific antibodies <Ang-2, VEGF> was measured according to the following test principle. HEK293-Tie2 clone 22 was stimulated with Ang-2 for 5 minutes in the absence or in the presence of Ang-2 antibody and P-Tie2 was quantified by a sandwich ELISA. Briefly, 2x105 cells of HEK293-Tie2 clone 22 were cultured per well overnight in a 96-well microtiter plate coated with poly-D-lysine in 100 µl of DMEM, 10% FCS, 500 mg / ml geneticin. The next day, a row of tetravalent bispecific antibody titers <Ang-2, VEGF> was prepared in a microtiter plate (concentrated 4 times, final volume of 75 μl / well, in duplicate) and mixed with 75 μl of one dilution of Ang-2 (R&D Systems, No. 623-AN) (3.2 mg / ml in the form of 4 times concentrated solution). Antibodies and Ang-2 were pre-incubated for 15 minutes at room temperature. 100 µl of the mixture was added to the HEK293-Tie2 clone 22 cells (preincubated for 5 minutes with 1 mM NaV3O4, Sigma No. S6508) and incubated for 5 minutes at 37 ° C. The cells were then washed with 200 µl of ice cold PBS + 1 mM NaV3O4 per well and lysed by the addition of 120 µl of lysis buffer (20 mM Tris, pH 8.0, 137 mM NaCl, 1% NP40 , 10% glycerol, 2 mM EDTA, 1 mM NaV3O4, 1 mM PMSF and 10 mg / ml aprotinin) per well on ice. The cells were lysed for 30 minutes at 4 ° C on a microtiter plate shaker and 100 µl of lysate was transferred directly to a p-Tie2 ELISA microtiter plate (R&D Systems, R&D No. DY990) without prior centrifugation and without determination of the total proteins. The amounts of P-Tie2 were quantified following the manufacturer's instructions and the IC50 values for the inhibition were determined using the XLfit4 analysis plug-in for Excel (one-dose dose response, model 205). The IC50 values can be compared within a given experiment, but may vary between experiments.

VEGF-induced HUVEC proliferation assay

HUVEC proliferation (endothelial cells of the human umbilical vein, Promocell No. C-12200) induced by VEGF was selected to measure the cellular function of tetravalent bispecific antibodies <Ang-2, VEGF>. Briefly, 5,000 HUVEC cells (reduced number of passes, pas5 passes) were incubated for each well of 96-well plate, in 100 μl of fasting medium (EMB-2, basal endothelial medium 2, Promocell No. C-22211, FCS 0.5%, penicillin / streptomycin) in a 96-well microtiter plate coated with collagen I BD Biocoat Collagen I (BD No. 354407/35640) overnight. Various concentrations of tetravalent bispecific antibody <Ang-2, VEGF> were mixed with VEGFrh (final concentration: 30 ng / ml, BD No. 354107) and pre-incubated for 15 minutes at room temperature. The mixture was then added to the HUVEC cells and incubated for 72 hours at 37 ° C, 5% CO2. On the day of the analysis, the plate was equilibrated at room temperature for 30 minutes and cell viability / proliferation was determined using the CellTiter-GloTM cell viability luminescent assay kit according to the manual (Promega, No. G7571 / 2/3). The luminescence was determined with a spectrophotometer.

Example 1

Production, expression, purification and characterization of a bispecific and tetravalent antibody that recognizes Ang5 2 and VEGF-A

In a first example, a tetravalent bispecific antibody was prepared without a linker between the respective antibody chains that recognized Ang-2 and VEGF-A, by fusion using a (G4S) 4 linker of a fusion of VH-CL domains against VEGF -A with the C-terminal end of the heavy chain of an antibody that

10 recognizes Ang-2 (SEC1 or a corresponding IgG1 allotype). In order to obtain the tetravalent bispecific antibody, said heavy chain construct was coexpressed with plasmids encoding the respective light chain of the Ang-2 antibody (SEC3) and a fusion of VL-CH1 domains that recognized VEGF-A (SEC2). The respective antibody scheme is provided in fig. 5.

The tetravalent bispecific antibody was generated as described in the general methods section, by classical molecular biology techniques, and was expressed transiently in HEK293F cells as indicated above. It was then purified from the supernatant by a combination of affinity chromatography with protein A and size exclusion chromatography. The product obtained was characterized for identity by mass spectrometry and analytical properties, such as purity,

20 via SDS-PAGE, monomer content and stability.

Expression

Purification

Title [μg / ml]

final product performance homogeneity (final product)

twenty-one

19.2 mg / L 95%

These data demonstrate that tetravalent bispecific antibody can be produced in good yield and that it is stable.

25 Next, Ang-2 and VEGF-A binding were studied, as well as simultaneous binding, by the ELISA and Biacore assays indicated above, and functional properties were analyzed, such as inhibition of Tie2 phosphorylation and inhibition. of VEGF-induced HUVEC proliferation, showing that the generated tetravalent bispecific antibody was capable of binding Ang-2 and VEGF-A and blocking its activities

30 simultaneously.

Example 2

Production, expression, purification and characterization of a bispecific and tetravalent antibody that recognizes Ang35 2 and VEGF-A

In a second example, a tetravalent bispecific antibody without linker between the respective antibody chains recognized by Ang-2 and VEGF-A was prepared by fusion with a connector (G4S) 4 of a fusion of VH-CL domains against VEGF- A with the N-terminal end of the heavy chain of an antibody that recognized

40 Ang-2 (SEC4 or a corresponding IgG1 allotype). In order to obtain the tetravalent bispecific antibody, said heavy chain construct was coexpressed with plasmids encoding the respective light chain of the Ang-2 antibody (SEC3) and a fusion of VL-CH1 domains that recognized VEGF-A (SEC2). The respective antibody scheme is provided in fig. 6.

The tetravalent bispecific antibody was generated as described in the general methods section by classical molecular biology techniques, and expressed transiently in HEK293F cells as indicated above. It was then purified from the supernatant by a combination of affinity chromatography with protein A and size exclusion chromatography. The product obtained was characterized for identity by mass spectrometry and analytical properties, such as purity, by SDS-PAGE,

50 monomer content and stability.

Expression

Purification

Title [μg / ml]

final product performance homogeneity (final product)

18

12.4 mg / L 95%

These data demonstrate that tetravalent bispecific antibody can be produced in good yield and that it is stable.

Next, Ang-2 and VEGF-A binding were studied, as well as simultaneous binding, by the ELISA and Biacore assays indicated above, and functional properties were analyzed, such as the inhibition of

Tie2 phosphorylation and inhibition of VEGF-induced HUVEC proliferation, showing that the generated tetravalent bispecific antibody was capable of binding Ang-2 and VEGF-A and blocking its activities simultaneously.

5 Example 3

Production, expression, purification and characterization of a bispecific and tetravalent antibody that recognizes Ang2 and VEGF-A

In a third example, a tetravalent bispecific antibody without a linker was prepared between the respective antibody chains recognized by Ang-2 and VEGF-A, by fusion via a connector (G4S) 4 of a Fab VH-CH1 domain against Ang -2 with the C-terminal end of the heavy chain of a CH1-CL exchange antibody that recognizes VEGF (SEC 5 or a corresponding IgG1 allotype). In order to obtain the tetravalent bispecific antibody, said heavy chain construct was coexpressed with plasmids encoding the chain

Respective light of the Ang-2 antibody (SEC3) and a fusion of VL-CH1 domains that recognized VEGF-A (SEC2). The respective antibody scheme is provided in fig. 7.

The tetravalent bispecific antibody was generated as described in the general methods section by classical molecular biology techniques, and expressed transiently in HEK293F cells as indicated.

20 previously. It was then purified from the supernatant by a combination of affinity chromatography with protein A and size exclusion chromatography. The product obtained was characterized for identity by mass spectrometry and analytical properties, such as purity, by SDS-PAGE, monomer content and stability.

Expression

Purification

Title [μg / ml]

final product performance homogeneity (final product)

8.5-9

1.8 to 4.1 mg / l 100%

25 These data demonstrate that tetravalent bispecific antibody can be produced in good yield and is stable.

Next, Ang-2 and VEGF-A binding were studied, as well as simultaneous binding, by ELISA assays.

30 and Biacore indicated above, and functional properties were analyzed, such as inhibition of Tie2 phosphorylation and inhibition of VEGF-induced HUVEC proliferation, showing that the tetravalent bispecific antibody generated was capable of binding Ang-2 and VEGF -A and to block their activities simultaneously.

35 LIST OF SEQUENCES

<110> F. Hoffmann-La Roche AG

<120> Bispecific antigen binding proteins

<130> 26145 WO

<150> EP09007857.7

<151> 2009-06-16

<160> 5

<170> PatentIn version 3.2

50 <210> 1

<211> 709

<212> PRT

<213> Artificial

55 <220>

<223> <Ang-2> of unmodified heavy chain with VH-CL domains of <VEGF> fused C-terminally of modified Fab fragment (CH1-CL exchange)

<400> 1

<210> 2

<211> 212 5 <212> PRT

<213> Artificial

<220>

<223> VL-CH1 domains of <VEGF> of modified Fab fragment (CH1-CL exchange)

<400> 2 <210> 3

<211> 215

<212> PRT

<213> Artificial <220>

<223> <Ang-2> of unmodified light chain

<400> 3 <210> 4

<211> 709 5 <212> PRT

<213> Artificial

<220>

<223> <Ang-2> of unmodified heavy chain with VH-CL domains of <VEGF> fused N-terminally of 10 modified Fab fragment (CH1-CL exchange)

<400> 4

<210> 5

<211> 709 5 <212> PRT

<213> Artificial

<220>

<223> <VEGF> modified heavy chain (CH1-CL exchange) with VH-CH1 domains of <Ang-2> fused 10 C-terminally of unmodified Fab fragment

<400> 5

Claims (9)

1. Bispecific antigen binding protein, comprising: 5 a) two light chains and two heavy chains of an antibody, which specifically bind to a first antigen and comprising two Fab fragments, b) two additional Fab fragments of an antibody that specifically binds to a second antigen, wherein said additional Fab fragments are fused, both by a linker peptide, with the heavy chains in a), where, in the Fab fragments are carry out the following modifications: i) in both Fab fragments of a), or in both Fab fragments of b), the variable domains VL and VH are replaced by one, and / or the constant domains CL and CH1 are replaced by one, ii) in both Fab fragments in a), the variable domains VL and VH are replaced one by another, and the constant domains CL and CH1 substitute each other, and 20 in both Fab fragments in b), the variable domains VL and VH are replaced by one, or the constant domains CL and CH1 are replaced by one, iii) in both Fab fragments in a) 25 the variable domains VL and VH are substituted for each other, or the constant domains CL and CH1 are mutually substituted, and in both Fab fragments in b), the variable domains VL and VH are substituted one for the other, and the constant domains CL and CH1 are replaced one by another, iv) in both Fab fragments in a), the variable domains VL and VH are replaced one by another, and in both Fab fragments in b), the constant domains CL and CH1 substitute each other, or v) in both Fab fragments in a), the constant domains CL and CH1 substitute each other, and in both Fab fragments in b), the variable domains VL and VH substitute each other. 2. Bispecific antigen-binding protein according to claim 1, wherein said additional Fab fragments are fused by a linker peptide with the C-terminal ends of the heavy chains in a) or with the N-terminal ends of the heavy chains in a). 3. Bispecific antigen binding protein according to claim 1 or 2, characterized in that the following modifications are carried out in the Fab fragments: i) in both Fab fragments of a), or in both Fab fragments of b), the variable domains VL and VH are substituted for one another, and / or the constant domains CL and CH1 are mutually substituted. 4. Bispecific antigen binding protein according to claim 3, characterized in that the following modifications are carried out in the Fab fragments: I) in both Fab fragments in a) the variable domains VL and VH are substituted for one another, and / or constant domains CL and CH1 are substituted for each other. 5. Bispecific antigen binding protein according to claim 4, characterized in that the following modifications are carried out in Fab 60 fragments: i) in both Fab fragments of a) the constant domains CL and CH1 are substituted for each other. 6. Bispecific antigen binding protein according to claim 3, characterized in that the following modifications are carried out in Fab 5 fragments: i) in both Fab fragments in b) the variable domains VL and VH are replaced by one, and / or the constant domains CL and CH1 substitute each other. 7. Bispecific antigen binding protein according to claim 6, characterized in that the following modifications are carried out in the Fab fragments: i) in both Fab fragments in b) 15 the constant domains CL and CH1 substitute each other. 8. Method for the preparation of a bispecific antigen binding protein according to claims 1 to 7, comprising the following steps: A) transforming a host cell with vectors comprising nucleic acid molecules encoding a bispecific antigen binding protein according to claims 1 to 7, b) culturing the host cell under conditions that allow the synthesis of said antigen-binding protein molecule, and c) recovering said antigen-binding protein molecule from said culture. 9. Host cell comprising the vectors according to claim 8. 10. Pharmaceutical composition comprising the bispecific antigen binding protein according to claims 1 to 7 and at least one pharmaceutically acceptable excipient. 11. Bispecific antigen binding protein according to claims 1 to 7, intended for the treatment of cancer. 12. Use of the bispecific antigen-binding protein according to claims 1 to 7, for the preparation of a medicament for the treatment of cancer.